{"id":"W2468149677","doi":"10.22215/etd/2012-09696","title":"Exploiting non-uniform query distributions in data structuring problems","year":2012,"lang":"en","type":"dissertation","venue":"","topic":"Data Management and Algorithms","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University; Canadian Heritage; Library and Archives Canada; Natural Sciences and Engineering Research Council of Canada","funders":"","keywords":"Structuring; Computer science; Information retrieval; Mathematics; Political science","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01353021,0.0009125086,0.002175313,0.001968932,0.001414759,0.003485635,0.003149039,0.002074055,0.002516493],"category_scores_gemma":[0.05521386,0.001302756,0.001271988,0.003245839,0.002536046,0.01093557,0.005674706,0.003332865,0.0005467252],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002353273,"about_ca_system_score_gemma":0.002392624,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001619062,"about_ca_topic_score_gemma":0.002389376,"domain_scores_codex":[0.9863854,0.007249466,0.0008950589,0.001746332,0.002724891,0.0009987756],"domain_scores_gemma":[0.9281865,0.05849744,0.002775438,0.007019607,0.002111076,0.001410005],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.001539539,0.0007405423,0.01016606,0.0005076879,0.0002596841,0.00028866,0.00117553,0.476131,0.008907548,0.219153,0.006747974,0.2743828],"study_design_scores_gemma":[0.00007850077,0.0001654095,0.0003528002,0.00001790804,0.00002743676,0.0001198591,0.0001342023,0.8721715,0.002063209,0.12367,0.001182824,0.00001630107],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"other","genre_scores_codex":[0.09561652,0.0007730508,0.8987586,0.001697532,0.0000399455,0.000111862,0.0001182638,0.0009130939,0.001971158],"genre_scores_gemma":[0.7103064,0.0007503292,0.2834936,0.0005419526,0.0003073653,0.000248771,0.0005192617,0.0005085879,0.003323651],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.01353021,"threshold_uncertainty_score":0.07155544,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04272130123280293,"score_gpt":0.2829957397456566,"score_spread":0.2402744385128536,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}